- 學位論文
Building Taiwan Law Ontology based on Automatic Legal Definition Extraction
Building Taiwan Law Ontology based on Automatic Legal Definition Extraction
摘要
隨著網際網路普及，現今各種專業領域知識也大量的數位化，藉由網路提供專業知識使用者可以快速地取得大量資訊。線上全國法規資料庫，也提供中華民國現行法規全文以供使用者檢索。但我國現行法規目前面臨的問題為，數量眾多且缺乏結構化的表達方式。為了準確且有效率地的提供法律知識給使用者(不論是法律專家、初學者或是使用者)，本研究提出以本體論及圖形化查詢介面，提供使用者快速查詢法律關鍵字之相關專業知識。本研究以中華民國現行法規為文本，利用自然語言處理工具與資料探勘方法，自動化擷取法律條文中之關鍵字及相關定義;接著，我們將關鍵字及相關定義建置符合本體論之知識結構，以自動化之方式建立台灣法律知識本體，並提供查詢介面讓使用者查詢專業知識。研究結果中，由法律條文中找出關鍵字及其定義之總數量為1114個。運用本體論之知識結構特性，使用者可以使用者查詢專業知識，並顯示其具關連性之法律關鍵字。本研究成果同時提供便利及圖形化之查詢介面，以利使用者查閱法律專業知識。
並列摘要
With the prevalence of the Internet, the professional knowledge in different domains can be accessed on the Internet. Through the Internet, users can acquire large amount of information. Laws & Regulations Database of The Republic of China, which is a governmental website, provides the current Chinese law text for users. But the problem is that we have a large number of law texts but lack of structured methods in writing. This thesis aims to provide an easy way for users to access professional knowledge and reduce time on searching and reading. In this thesis, we use current Chinese law text from Taiwan law database as data source. We also utilize the Natural language processing tool and the data mining mechanisms to automatically extract legal keywords and mine the definition of legal keywords. In the next step, we construct Taiwan law ontology with the legal keywords and relative definitions. We also provide the service on the web such that users can access the legal information on the Internet. In the result, we get 1114 legal keywords with definitions. With the characteristics of ontology, users can view not only one keyword and its information, but also the associated keywords. For the web interface, we provide graphical and text interfaces for users to browse legal knowledge.